Jurnal Anak

Paediatrica IndonesianaVOLUME 50 NUMBER 2March

Original Article

Paediatr Indones, Vol. 50, No. 2, March 201073

Comparison of bone age in small-for-gestational-age children vs appropriate-for-getational-age

children

Edward Lionardus, Sjarif Hidajat Effendi, Djatnika Setiabudi

AbstractBackground$ERXWVPDOOIRUJHVWDWLRQDODJHFKLOGUHQDUHin higher risk for having linear growth retardation due to growth KRUPRQHLQVXOLQ OLNH JURZWK IDFWRU D[LV GHIHFW *+,*)which causes bone age delay.Objectives 7RFRPSDUHERQHDJH LQPRQWKROGFKLOGUHQborn small-for-gestational-age (SGA) to that in children born appropriate-for-gestational-age (AGA).Methods A cross-sectional study was conducted in Hasan Sadikin General Hospital, Bandung, from January to April 2009.6XEMHFWVFRQVLVWHGRI KHDOWK\FKLOGUHQRIPRQWKVROGchildren born at term, SGA, 25 children born at term, AGA). We compared the appropriateness and delay of bone age between the two groups. Results 0HDQERQH DJH LQ WKH6*$JURXSZDV 6'PRQWKV DQG LQ WKH$*$ JURXSZDV 6' PRQWKV(P 0HDQERQH DJHGHILFLWZDV PRQWKV LQWKH6*$JURXSDQG 6'PRQWKV LQ WKH$*$JURXS(P 7KHSUHYDOHQFHUDWLRZDV&,Bone age delay was found to be higher in children born SGA than WKDWLQFKLOGUHQRIWKHRWKHUJURXSYV2QWKHFRQWUDU\DSSURSULDWHERQHDJHZDVIRXQGPRUHLQFKLOGUHQERUQ$*$vs 2) (P=0.002).Conclusion%RQHDJHGHOD\LQPRQWKVROGFKLOGUHQERUQsmall-for-gestational-age was found to be higher than in those born appropriate-for-gestational-age. [Paediatr Indones. 2010;50:73-9].

Keywords: Small-for-gestational-age, appropriate-for-gestational-age, bone age, growth retardation

)URP'HSDUWPHQW RI &KLOG+HDOWK0HGLFDO 6FKRRO 3DGMDGMDUDQUniversity, Hasan Sadikin General Hospital, Bandung, Indonesia.

Reprint request to: Edward Lionardus, MD, Department of Child Health, 0HGLFDO6FKRRO3DGMDGMDUDQ8QLYHUVLW\+DVDQ6DGLNLQ*HQHUDO+RVSLWDO-O3DVWHXU1R%DQGXQJ,QGRQHVLD7HO)D[([email protected]

Small-for-gestational age (SGA) refers to babies who have birth weight and/or birth OHQJWK OHVV WKDQ th percentile according to Lubchenco gestational age curve.$FFRUGLQJWRSUHVHQWGDWDWKHUHZHUHDERXWof SGA babies.Ministry of Health of Indonesia stated WKDWLQWKHUHZHUHEDELHVERUQDV6*$LQIndonesia.5 SGA babies will have intrauterine growth retardation, which impacts GH-IGF axis (growth hormone-insulin like growth factor) that delays bone maturity. Resulting in growth spurt impairment in their childhood related to short stature.

Most SGA children will do their growth catch up DQGJHWWKHLUILQDOKHLJKW!6'WKLVSURFHVVVWDUWVin the first year of life and will be completed in 2 years of age.$ERXWRI6*$FKLOGUHQZLOOKDYHWKHLUKHLJKW6'Study by Hediger et al found WKDW6*$FKLOGUHQZLOOKDYHKHLJKWGHILFLWDWyears old. Short stature children born SGA who do not catch their growth spurt at 2-3 years old have to be managed by a pediatric endocrinologist.2

Edward Lionardus et al: Bone age in small-for-gestational-age children

74 Paediatr Indones, Vol. 50, No. 2, March 2010

The relationship between the etiology of intra-uterine growth retardation and postnatal growth pattern is still unexplainable, but it is thought to be due to GHIHFWLQ*+,*)D[LV7KLVLVVXSSRUWHGE\HYLGHQFHthat early mechanism of growth spurt is hypersecretion of growth hormone (GH). Studies show low GHsecretion and insulin growth IDFWRU ,*) VHUXPconcentration in children with failure to catch up,

and GH replacement therapy in short stature children will accelerate growth spurt and significantly increasebody height. The aim of this study was to determine and WRFRPSDUHERQHDJHLQFKLOGUHQDJHGPRQWKVERUQSGA to that in children born AGA.

Methods

In this cross sectional study we included babies born at term, appropriate-for-gestational-age and those born small-for-gestational-age (AGA and SGA) in+DVDQ6DGLNLQ*HQHUDO+RVSLWDOIURP-DQXDU\WR

'HFHPEHU$W WKH VWDUWRI WKH VWXG\ WKHEDELHVZHUHPRQWKVROG:HH[FOXGHGFKLOGUHQwith chronic disease (tuberculosis, hepatic cirrhosis,nephrotic syndrome, and malignancy), severe PDOQXWULWLRQDQGPDMRUFRQJHQLWDODQRPDO\3DUHQWDOLQIRUPHGFRQVHQWZDVREWDLQHGIURPDOOVXEMHFWV

:HHVWLPDWHGWKHQXPEHURIUHTXLUHGVXEMHFWVE\ PHDQV RI SORW VWXG\ VXEMHFWV SHU JURXSZHUHQHHGHG %RQHDJHZDVPHDVXUHGXVLQJ;UD\examinations of the left hand by two radiologists. The results were compared to standard presented inGreulich-Pyle atlas. Bone ages result was described in months and was considered appropriate if the resultswithin 3 months.

To analyze parents characteristics, nutritional status, and bone age ratio to chronologic age data, weused x2WHVW7RDQDO\]HVXEMHFWVFKDUDFWHULVWLFVGDWDwe used t-test. The prevalence ratios were measured. Appropriateness of bone age results between two radiologists were tested using coefficient of agreementKappa. To measure Kappa index (.), we used 2

Figure 1. Ilustration of birth in Hasan Sadikin Hospital, Bandung, 2006 and methods todetermine study subjects

SGA = small-for-gestational-ageAGA = appropriate-for-gestational-ageLGA = large-for-gestational-ageIUFD = intrauterine fetal deathTBC = tuberculosis



x2 table. Comparison of bone age results by two radiologists was analyzed using Wilcoxon test and appropriateness of bone age results by two radiologists was analyzed using McNemar x2WHVW3YDOXHRIwas considered statistically significant. We use SPSS YHUVLRQIRU:LQGRZV6366LQF&KLFDJRIllinois, USA for all analyses.

This study had been approved by Ethical Com-PLWWHHRI0HGLFDO)DFXOW\RI3DGMDGMDUDQ8QLYHUVLW\Hasan Sadikin General Hospital, Bandung.

Results

7KHUH ZHUH EDELHV ERUQ LQ+DVDQ 6DGLNLQ*HQHUDO+RVSLWDOIURP-DQXDU\WR'HFHPEHUFRQVLVWHGRIFKLOGUHQERUQ6*$FKLOGUHQERUQ$*$DQGFDVHVRILQWUDXWHULQHIHWDOGHDWK,8)')URPWKH6*$FKLOGUHQWKHUHZHUHFKLOGUHQERUQDVWHUPDQGDVSUHWHUPEDELHV2XWRIFKLOGUHQERUQ6*$RIWKHPGLHGZHUHenrolled, and 38 children could not be traced. From the 38 enrolled children, there were four children excluded because of tuberculosis disease, so in the EHJLQQLQJZHKDGFKLOGUHQHQUROOHGWRWKHVWXG\DQGRI WKHPZHUH UDQGRPO\ VHOHFWHG ER\VJLUOV7HUP$*$FKLOGUHQZHUHVHOHFWHGDVWKHcontrol group with well-matched age and sex.

*HQHUDOFKDUDFWHULVWLFVRIWKHVXEMHFWVLQFOXGHGsex, birth weight, birth length, gestational age, chronological age, body weight, body height, nutritional status, parents educational level, and family income (Table 1).

Birth weight and birth length of the SGA group were smaller compared to those of the AGA group, and VWDWLVWLFDOO\VLJQLILFDQW30HDQERG\ZHLJKWRIWKH$*$JURXSZDVNJFRPSDUHGWRWKH6*$JURXSNJDQGDOVRVWDWLVWLFDOO\VLJQLILFDQWThere were more severely stunted children in the SGA JURXSFRPSDUHGWRWKH$*$JURXSYVEXWthe nutritional status was not statistically significant according to the weight/height, height/age curve, and weight/age. There were no significant statistically differences between fathers and mothers educational level, and family income [data not shown].

In Table 2, we can see that both radiologists results were not significantly different for either SGA or AGA group.

Table 3 shows that bone age measurement EHWZHHQWZRUDGLRORJLVWVFRQVLVWHGRIZHOOPDWFKHGUHVXOWVQLQHZHUHDSSURSULDWHDQGZHUHGHOD\HGand seven ill-matched results (two cases determined DVGHOD\HGE\$EXWDV DSSURSULDWHRQHVE\8DQGfive case determined as appropriate bone age by A, but as delayed bone age by U). A kappa index was

Table 1. General characteristics

Subjects characteristicsSGA AGA

n = 25 n = 25Sex

Birth weight, mean (SD) g Birth length, mean (SD) cmWeight, mean (SD) kgHeight, mean (SD) cm

Nutritional status W/H

OverweightMedianWasted

H/AMedianStuntedSeverely stunted

W/AMedianUnderweight

11/142,209 (248)46.0 (2.6)

11.04 1.23)84.3 (4.6)

0169

1375

178

11/143,119 (304)49.1 (1.4) 11.9 (1.5)86.3 (4.9)

1195

1361

205

Note: # = t-test; * = x2 test; SD =standard deviation; W=weight; H=height; A=age

Table 2. Comparison of bone age measurement in 24-36 month old children born SGA and AGA between two radiologists (n=50)

VariableRadiologist

5KIPKECPEGA UBone age, mean (SD) mo

MedianInterval

&GEKVOGCP 5&mo

MedianInterval

23.24 (7.76)24

3-36

-8.02 (6.56)-9

-27 s/d 4

22.74 (7.80)18

6-36

-8.50 (6.46)-9

-19 s/d 5

Zw = 0.675P = 0.50

Zw = 0.638P = 0.524

Note: Zw = Wilcoxon test

Table 3. Agreement of the bone age result of 24-36 months old children born SGA and AGA between two radiologists

Bone AgeU

NDelay Appropriate

ADelay

Appropriate

34

5

2

9

36

14Number 39 11 50

Note: x2 McNemar; P = 0.453; Kappa Index = 0.628


76Paediatr Indones, Vol. 50, No. 2, March 2010

ZKLFKPHDQWZHOOPDWFKHG)OHLVVRUVXEVWDQWLDO(Landis and Koch).

5DGLRORJLFERQHDJHPHDVXUHPHQWRIWKHVXEMHFWVwas compared to Greulich and Pyle atlas. In Table 4,we see significant differences between means of bone age of SGA and AGA. Significant differences were also found in means of bone age deficit compared to chronological age between the SGA and AGA.

Table 5 shows that in the SGA group there were 23 (92%) children with delayed bone age compared WRFKLOGUHQLQWKH$*$JURXS,QWKH6*$group, there were only two children with appropriate ERQHDJHFRPSDUHGWRFKLOGUHQLQWKH$*$JURXS&RPSDULVRQRIERQHDJHLQPRQWKVROGFKLOGUHQbetween those born SGA and those of AGA group was statistically significant (P=.002).

Discussion

The incidence of newborn with SGA in our study ZDV TXLWH VLPLODU WR SUHYLRXVGDWD WKDWWKHUHZHUHEDELHV ERUQ6*$ the overall LQFLGHQFHRI6*$LQ,QGRQHVLDLV5 The mean birth weight and length of the SGA group were smaller compared to those of the AGA group. Mean birth ZHLJKWRIWKH6*$JURXSZDV6'JUDPDQGit was statistically significant smaller compared to the $*$JURXS6'JUDP30HDQbody length of the SGA group was also statistically significant shorter compared to the AGA group.

There was no statistical differences in the gestational age between SGA and AGA group. These data show that children born AGA have heavier birth weight and longer birth length compared to SGA children. SGA was defined as children who were born with birth ZHLJKWELUWKOHQJWKOHVVWKDQth percentile according to the population data for identical gestational age according to the Lubchenco curve.

The mean chronologic age in the SGA and AGA group were not significantly different. There was different mean body height of the SGA group compared to the AGA group. This was similar to the results of the study conducted by Hediger et al.

But with different interval. Study of Strauss and Dietz25VKRZHGDGHILFLWDWWKHDJHRI\HDUVFrom the previous studies, we can see that the older the children, the bigger the height deficit in the SGA group compared to the AGA group, which finally resulted in short stature.

In the SGA group there were five children with KHLJKWDJH6'VHYHUHO\VWXQWHGFRPSDUHGWRWKH$*$JURXSRQHFKLOG%DELHVERUQ6*$DUHLQKLJKHUrisk for having growth retardation (short stature). In WKLVVWXG\ZHIRXQGKLJKHULQFLGHQFHFRPSDUHGto the study done by Karlberg et al dan Leger et al

ZKLFK IRXQG WKDW DERXW RI 6*$FKLOGUHQZLOO KDYH WKHLU KHLJKW 6'1XWULWLRQDO VWDWXVexaminations showed no statistical differences, either using the weight/height, height/age, or weight/age.

Bone age were examined by two radiologists using standardized Greulich and Pyle atlas.

The results of the examinations between the two radiologists were tested using the coefficient agreement .DSSD7KH.DSSD LQGH[ZDV DQGLW ZDV QRW VWDWLVWLFDOO\ GLIIHUHQW 3 ZKLFKmeans that the two radiologists were well-matched (according to Fleiss) or substantial (according to Landis and Koch).30 Comparison between both radiologists showed no significant differences on bone age variable (P=0.50) as well as on bone age deficit variable (P=0.52). The bone age examinations showed a reliable and consistent data.

This study showed that comparison of bone age GHOD\ LQ PRQWKROG FKLOGUHQERUQ6*$DQGAGA was similar to previous prediction. Bone age of the SGA and AGA group showed significant statistical differences: there was more children with bone age delay in the SGA group compared to the AGA group.

Table 4.%QORCTKUQPQHDQPGCIGCPFDQPGCIGFGEKVDGVYGGPtwo groups

ResultsGroup

PvalueSGA

n = 25AGA

n = 25Chronological age, mean (SD) moBone age, mean (SD) mo$QPGCIGFGEKVOGCP5&OQ

31.0 (3.8)20.8 (7.7)-10.5 (6.5)

31.0 (3.8)25.7 (7.1)-5.5 (5.7)

.732#.022*.009*

Note: # = t-test; * = Mann-Whitney test; SD = standard deviation

Table 5. Comparison of bone age in 24-36 months old SGA and AGA children

Delay/ AppropriateGroup

P value*SGAn = 25

AGAn = 25

DelayAppropriate

232

1312

0.002

Note: * = x2; Prevalence ratio = 1.77 (95% CI:1.192.62)



Comparison between chronologic age and bone age of the SGA and AGA children showed significant difference. More children in the SGA group had delayed bone age compared to those in the AGA group YVDQGPRUHFKLOGUHQLQWKH$*$JURXSKDGappropriate bone age compared to those in the SGA JURXSYV7KH35ZDV&,WRZKLFKPHDQVWKDWFKLOGUHQLQWKH6*$JURXSKDYHWLPHVKLJKHUULVNIRUKDYLQJGHOD\HGERQHage. SGA children will have delayed bone age due to *+,*)D[LVGHIHFWWKDWZLOOGHFUHDVH*+,*)secretion level.

Bone maturity is one of growth process and influenced by many factors such as genetics, hormonal, chronic diseases, nutrition, and social-economic status. Confounding factors in this study were social-economic status (parents educational level and family income), which showed no significant differences.

Genetic factor is one of the factors that influences bone maturity. Study of Arends et al33

VKRZHGWKDW6*$EDELHVZLWK,*)DOOHOHZLOOEHshorter than those without. Research by Abuzzahab et al35 also showed a correlation between pre- and post QDWDOJURZWKUHVWULFWLRQZLWKJHQH,*)5PXWDWLRQIn this research, they found insulin-like growth factor releasing factor (IGF-RF) gene mutation in SGA FKLOGUHQWKDWFKDQJHVWKHDUUDQJHRI$UJ*OQLQRQHDOOHOHDQG/\V$VQLQRWKHUDOOHOHZLWKSRLQWmutation CGATGA (Arg59 stop) that decrease the DPRXQWRI,*)UHFHSWRU

Linear growth can be measured using the body height and bone age. There was no significant GLIIHUHQFHV LQ ERG\ KHLJKWPHDVXUHPHQW LQ months old children, although there was height deficit -0.3 SD in the SGA group compared to the AGA group. Growth measurement using bone age is a better tool compared to body height measurement because body height measurement only describes actual mean time growth of the children, while bone age measurement can be used to determine growth status and to predict prognosis and final body height.

There were several limitations of this study. )LUVWZHUHFUXLWHGWKH$*$DQG6*$VXEMHFWVIURPmedical record. Second, we did not classify SGA into symmetric and asymmetric, so we could not determine which SGA group that has higher risk

for having bone age delay. In this study, we could RQO\FRQFOXGHWKDW6*$FKLOGUHQKDYHWLPHVhigher risk for having bone age delay compared to the AGA children. Third, we did not measure the *+ DQG ,*) OHYHO RI WKH VXEMHFWV:H VXJJHVWfurther research on bone age delay that also classified WKH 6*$ JURXS LQFOXGLQJ *+ DQG ,*) OHYHOmeasurement.

We concluded that there is bone age delay in PRQWKVROGFKLOGUHQPRUHLQWKRVHERUQ6*$compared to those born AGA.

References

$ONDOD\$/*UDKDP -U -03RPHUDQFH -- (YDOXDWLRQ RIneonates born with intrauterine growth retardation: review DQGSUDFWLFHJXLGHOLQHV-3HULQDWRO

2. Lee PA, Chernausek SD, Hokken-Koelega ACS, Czernichow P. International small for gestational age advisory board consensus development conference statement: management RIVKRUWFKLOGUHQERUQVPDOOIRUJHVWDWLRQDODJH$SULOth-2FWREHUst3HGLDWULFV

6KDOLWLQ6/HEHQWKDO


78Paediatr Indones, Vol. 50, No. 2, March 2010

3HGLDWU5HV $OEHUWVVRQ:LNODQG..DUOEHUJ - 3RVWQDWDO JURZWK RI

children born small for gestational age. Acta Pediatr Supp.

2QJ./$KPHG0/ (PPHW 30 3UHHFH0$'XQJHUDB. Association between post-natal catch-up growth and obesity in childhood: prospective cohort study. Br Med J.

.DUOEHUJ-$OEHUWVRQ:LNODQG.*URZWKLQIXOOWHUPVPDOOfor-gestational-age infants: from birth to final height. Pediatr 5HV

/HJHU - /HY\0DUFKDO&%ORFK - 3LQHW$&KHYHQQH'Porquet D, et al. Reduced final height and indications for insulin resistance in 20 year olds born small for gestational DJHUHJLRQDOFRKRUWVWXG\%U0HG-

'H=HJKHU))UDQFRLV,9DQ+HOYRULW09DQGHU%HUJKH*Clinical review 89: small as fetus and short as child: from endogenous to exogenous growth hormone. J Clin Endocrinol 0HWDE

'H:DDO:-+RNNHQ.RHOHJD$&66WLMQHQ7GH0XLQFNKeizer-Schrama SMPE, Drop SLS, and the Dutch group on growth hormone. Endogenous and stimulated GH secretion, urinary GH excretion, and plasma IGF-I and IGF-II levels in prepubertal children with short stature after intrauterine JURZWKUHWDUGDWLRQ&OLQ(QGRFULQRO

$FNODQG)06WDQKRSH5(\UH&+DPLOO*-RQHV-3UHHFHMA. Physiological growth hormone secretion in children with short stature and intra-uterine growth retardation. Horm Res.

$OEHUWVRQ:LNODQG. IRU WKH 6ZHGLVK 3DHGLDWULF 6WXG\Group for Growth Hormone Treatment. Growth hormone secretion and growth hormone treatment in children with intrauterine growth retardation. Acta Paediatr Scand Suppl.

5RFKLFFLROL37DXEHU00RLVDQ93LHQNRZVNL&,QYHVWLJDWLRQof growth hormone secretion in patients with intrauterine growth UHWDUGDWLRQ$FWD3DHGLDWU6FDQG6XSSO

$UHQGV1-7%RRQVWUD9++RNNHQ.RHOHJD$&6+HDGcircumference and body proportions before and during growth hormone treatment in short children who were born VPDOOIRUJHVWDWLRQDODJH3HGLDWULFV

20. Clayton PE, Cianfarani S, Czernichow P, Johannsson G, Rapaport R, Rogol A. Consensus statement: management of the child born small for gestational age through to adulthood: a consensus statement of the international societies of pediatric endocrinology and the growth hormone research VRFLHW\-&OLQ(QGRFULQRO0HWDE

7KXUHHQ 3- $QGHUVRQ06+D\:: 7KH VPDOOIRU

JHVWDWLRQDODJHLQIDQW1HRUHYLHZVH22. Wilkinson AR, Charlton VE, Phibbs RH, Amiel-Tison

C. Examination of the newborn infant. In: Rudolph CD, 5XGROSK$0HGLWRUV5XGROSKVSHGLDWULFVst edition. New



35. Abuzzahab MJ, Schneider A, Goddard A, Grigorescu F, Lautier C, Keller E, et al. IGF-I receptor mutations resulting in intrauterine and postnatal growth retardation. N Engl J 0HG

%URRN&**URZWKDVVHVVPHQWSXUSRVHDQGLQWHUSUHWDWLRQIn: Brook CG, Hindmarsh PC, editors. Clinical pediatric HQGRFULQRORJ\ th edition. London: Blackwell Science &RPSDQ\S

Documents

Jurnal Anak